1[
2{ type: install
3  message: <<EOM
4You installed LLaMA-cpp: Facebook's LLaMA model runner.
5
6In order to experience LLaMA-cpp please download some
7AI model in the GGUF format, for example from huggingface.com,
8run the script below, and open localhost:9011 in your browser
9to communicate with this AI model.
10
11$ llama-server -m $MODEL \
12  --host 0.0.0.0 \
13  --port 9011 \
14  -ngl 15
15
16or
17
18you can add the following lines to /etc/rc.conf,
19start the llama-server service,
20and navigate to http://localhost:8080:
21> llama_server_enable=YES
22> llama_server_model=/path/to/models/llama-2-7b-chat.Q4_K_M.gguf
23> llama_server_args="--device Vulkan0 -ngl 27"
24
25EOM
26}
27]
28